Intuit

IThe business

Intuit sells software and services that handle money chores. TurboTax files US tax returns. QuickBooks keeps the books of small and mid-sized businesses. Credit Karma shows people credit scores and sends them to lenders. Mailchimp does email marketing. Intuit says it serves about 93 million customers, and the company had 18,600 employees on 31 July 2026.² About 8% of revenue is outside the United States.²

The business has two segments. Global Business Solutions, which is 60% of revenue, is QuickBooks, payroll, payments and small-business loans. Consumer, 40%, is TurboTax, Credit Karma and a small professional tax line.² Each segment earns a very high profit before the costs the company keeps at the platform level. Those costs are large: $10.3B in fiscal 2026, almost half of revenue, and they include stock pay, hosting, support experts and amortization of acquired assets.²

In ten years revenue has grown from $5.2B to $21.4B, a compound rate of 17% a year, and GAAP operating income from $1.4B to $5.9B, 17% a year C (calculated).⁹² Part of that was bought. Credit Karma arrived in December 2020 for $7.2B and Mailchimp in November 2021 for $12.0B.⁸⁷ The rest was price, mix and new products.

Tax: seasonal, profitable and under attack at the bottom

TurboTax earned $5.3B in fiscal 2026, 7% more than a year before, on 39.0M federal returns, which was 2% fewer.³ Revenue rose while units fell because each customer paid more. Revenue per online filer rose about 11% on the company’s own figure, as more people chose to pay for help from an expert and for faster access to their refund.⁴ TurboTax Live, the version with a human expert, grew 37% and is now 53% of TurboTax revenue.³

The weak spot is the simple, do-it-yourself filer. Goodarzi told investors in May that the company faced pressure from filers earning under $50,000 and said, “We lost on price.”⁴ In his words the do-it-yourself segment is a $5B market, or 12% of the total TurboTax opportunity, and the assisted segment is $37B.⁴ The company said that about 7M filers paid nothing this year, down from 8M, and that total online units would fall about 2% while paying units rose 2%.¹¹ So the fall in units is mostly people who were not paying anyway. The price response matters all the same. At the August call, according to a summary, management said it would make the entry-level product cheaper and accept lower revenue per user at the start.¹² I treat that summary as secondary and read the prepared remarks as the primary record.

Tax is also seasonal. Most TurboTax revenue arrives between November and April, so the second and third quarters carry the year.² Intuit also said that total IRS filers were expected to fall about 30 basis points this season, which it put at about 2M returns short of what the market expected.⁴ I can’t verify that estimate.

Small business: price, mix and money

QuickBooks Online Accounting revenue grew 23% to $5.1B. Intuit credits “higher effective prices, customer growth, and mix shift.”² The detail matters. Paying customers across the Online Ecosystem rose only 3%, while revenue per customer rose 15%.² Most of the growth is price and better-priced plans, not a bigger base. That is how a mature software product grows, and it has a limit.

The faster parts are payments and loans. Money revenue rose $434M, of which payments were $257M and QuickBooks Capital, the loan product, $177M.² Total online payment volume, including bill pay, grew 30% in the third quarter.⁴ The mid-market products, QuickBooks Advanced and Intuit Enterprise Suite, grew about 38%.⁴ These are the “Big Bets” that management points to: assisted tax, money and mid-market. Together they grew 34% and made up 30% of revenue.³

Credit Karma and the lenders behind it

Credit Karma earned $2.6B, up 20%. Of that increase of $441M, personal loans were $235M, credit cards $101M and insurance $85M.² Credit Karma is paid when a lender approves a customer, so its revenue depends on how much banks want to lend. It grows when credit is loose and shrinks when it tightens. I have no Credit Karma history from before the 2020 purchase to test that.

What changed in 2026

The stock fell 11% on 3 February and 20% on 21 May, after the third-quarter report, from $383.93 to $307.07 C (calculated).¹ That was the largest one-day fall in my ten-year file of closes C (calculated).¹ The company said it would cut 17% of its full-time staff, wind down offices in Reno and Woodland Hills, and invest less in Mailchimp.¹³ It took a $293M charge in the fourth quarter and expects about $315M in total.² An earlier cut had cost $223M in fiscal 2024.² In July and August two shareholder class actions were filed that allege misleading statements in 2025 and 2026.² Intuit says it will defend them.

It then guided to 9% to 10% revenue growth for fiscal 2027, slower than any of the ten years in my table, and to GAAP operating income up 26% to 27%.³ From 1 August 2026 it also counts stock pay in its non-GAAP profit.³ I come back to that in Financials, because it changes what “earnings” means here.

IIThe moat

The protection has four layers, and they are not equally strong. Taxes and books are high-stakes. A wrong number costs money and a penalty, and the filer remembers who gave it. Intuit says customers spend at least seven times more on experts than on software, and that confidence is what they buy.⁴ That is a company claim, but it fits how TurboTax Live took over half the product.

Evidence it is holding

Revenue per online filer rose about 11% in a season with fewer filers.⁴ Assisted tax grew more than 30% and is now more than half of TurboTax.⁴³ The IRS’s own free filing system was suspended.² H&R Block, the other large name in tax, grew revenue 4.9% to $3.95B and its do-it-yourself line was flat at $385M.⁵ In QuickBooks the mid-market tiers grew about 38% in the third quarter.⁴ None of that looks like a business losing its customers.

Evidence it is narrowing

Three things point the other way. The first is the company’s own admission on price. The second is the customer count: the 10-K says Intuit serves about 93 million customers, against about 100 million a year earlier, though the definitions may differ.²¹⁵ The third is distribution. Intuit has struck a deal to put its apps inside ChatGPT, reported at more than $100M.¹⁴ That can bring customers, and it also teaches people to start in someone else’s product. The 10-K lists “platform companies and providers of general purpose AI technologies” among its competitors.²

The government is a fourth. The 10-K says the IRS free direct filing system “was suspended” and warns that its proponents continue to advocate for it.² A federal tax tool that returns would hit the simplest filers first, the same group Intuit says it lost on price.

The AI question

This is why the stock fell, so I want to say plainly what I can and can’t judge. A general AI assistant can now read a W-2, ask questions and fill a form. That could take the front of the tax business: the first click, the easy return, the cheap answer. What it can’t easily do is take liability for the answer, file with all 50 states, and be audited. Intuit’s bet is that people want a done-for-you service with a human in it. The results so far fit that bet: revenue per filer is up, and assisted is growing 30%+. But one tax season is not a test, and the first filers lost were the ones with the least to lose. I can’t judge how fast a free assistant becomes good enough for a complex return, and I show a range for that.

IIIThe competitive advantage

Intuit’s edge is that it sits at both ends of a customer’s money life. A person files with TurboTax, gets a refund advance, sees credit offers on Credit Karma, and may run a business on QuickBooks. Management says average revenue is about 30% higher for customers who use both TurboTax and Credit Karma.⁴ A rival with only one of these products has to find the rest elsewhere.

Where Intuit wins

It wins on breadth. No rival I looked at has a tax product, a small-business ledger, payments, payroll and a credit marketplace in one account. It wins on the accountant side, where its software is the habit of about a million professionals.⁴ And it wins on profit: both segments earn above 70% before platform costs.²

Where it is weaker

It is weaker at the low end of tax and at the front door. “We lost on price” is the company’s phrase.⁴ It is weaker at the bottom of small-business accounting, where Intuit itself launched QuickBooks Free and QuickBooks Lite this year to meet cheaper rivals.⁴ And it is weaker on the AI side in one way that no filing can show: the AI assistant, not Intuit, decides which product to send a user to.

Mailchimp is the clear miss. Intuit paid $12.0B for it in November 2021, and in eight months it brought in $762M of revenue.⁷ The fiscal 2027 guide for Mailchimp is $1,256M to $1,266M, down 1% to flat, which implies about $1.27B in fiscal 2026 C (calculated).³ That first eight months was a pace of about $1.14B a year, so revenue is about 11% higher five years on, and flat now C (calculated). The company cut investment in it in May.¹³ I can’t see the segment’s profit, as Intuit has not shown it, so I can’t compute a return on the $12.0B. Revenue alone says it was an expensive purchase.

IVManagement

Sasan Goodarzi, 58, has been chief executive since January 2019 and chairman since January 2026. He joined Intuit in 2004, left for ten months in 2010, and returned in 2011. He ran the small-business group and the consumer tax group before the top job.²⁶ Scott Cook, 73, the co-founder, is still on the board and owns 6.16M shares, 2.2% of the company, or about $2.1B at today’s price C (calculated).⁶¹

How they are paid, and what that pays them to do

Goodarzi’s reported pay for fiscal 2025 was $36.8M, which is 152 times the median employee’s $242,190.⁶ His fiscal 2025 salary was $1.3M, and it is $1.4M for fiscal 2026. His bonus target is 200% of salary, and for fiscal 2025 he was paid 100% of target, $2.6M. The fiscal 2025 equity grant targeted $34.6M, split into 50% performance shares, 25% restricted stock units and 25% options.⁶ The company calls 96% of his pay performance-based.⁶

Two parts of that plan tell me what it rewards. The bonus pool is set by revenue and non-GAAP operating income. Until August 2026 that profit figure left out stock pay, which was $2.0B a year.⁶³ Management has now changed it, and the new guide includes $2,020M of stock pay.³ I take that as a good sign, and also as an admission that the old measure flattered. The performance shares pay on Intuit’s return against 43 other technology companies over one, two and three years, with target at the 60th percentile, and they are capped at 100% if the absolute return is negative.⁶ With the stock 58% below its high, I expect those grants to pay little E (estimate).¹ That is a real test of alignment and I count it in their favour.

The pay plan rewards growth in revenue and operating income and a good stock return against peers. It does not reward a return on the money spent on acquisitions, which is the gap I come to next. One more point: the committee paid 100% of target in fiscal 2025, though the formula gave 117.8%, because the company missed some non-financial goals.⁶ That shows some restraint.

How much of their own money is in

The proxy’s ownership table, dated 31 October 2025, shows Goodarzi with 52,545 shares through a trust, about $17.9M at today’s price, plus 231,567 shares he could get from options and vested units C (calculated).⁶¹ The company requires ten times salary in stock, which is $14M.⁶ So he meets the rule, and most of what he owns came from pay, not from purchases.

The Form 4 filings since August 2025 show what insiders did with the stock. Goodarzi sold 41,000 shares on 7 January 2026 at about $650, for $26.7M. Cook sold 494,894 shares between August and December 2025, for $327.9M at about $663 a share. Aujla sold 3,644 shares for $2.4M.¹⁷ In March 2026 the company said that its founder and executive team had ended all of their stock-sale plans.¹⁸ In the whole period I found two open-market purchases, both by a director, Vasant Prabhu: 1,750 shares at about $310 between 22 and 26 May, just after the fall.¹⁷ The sales were at about twice today’s price, and no one in management has put new money in since the fall. I don’t read them as a signal of what these people think of the stock now, since a fall of this size was not in view then.

What they did with capital over a full cycle

Over ten years Intuit earned about $28B of owner cash and paid out $25B of it in buybacks and dividends C (calculated).⁹ It also spent $9.1B of cash and issued 20.7M shares to buy Credit Karma and Mailchimp.⁸⁷ To fill the gap it added about $7B of debt.²¹⁰ Even so, diluted shares rose from 261M to 277M, because the two deals issued stock and employees receive about $2B of it a year.⁹

The two deals had different outcomes. Credit Karma, bought for $7.2B in purchase price ($8.1B with assumed awards and restricted shares), now makes $2.6B a year of revenue and grew 20% last year.⁸² That looks like a good purchase. Mailchimp, bought for $12.0B, makes about $1.27B, is guided down, and had its investment cut.⁷³¹³ I count one clear success and one clear miss on $19.2B. The buyback record is also mixed. The company bought 4.3M shares in fiscal 2025 at about $650, and 13.4M in fiscal 2026 at about $410.¹⁵² The price is now 27% below the average of the two years C (calculated).¹

Where I come down on management: a capable operator who has built a very profitable franchise, a company that now tells investors the hard thing (“we lost on price”), and an acquisition record that I would score at one clear hit and one miss. The say-on-pay vote in January was 93% in favour. About 4% of votes were against Goodarzi’s re-election and 6% against Deborah Liu, the highest for any director, and 8% were against the auditor.¹⁹ Those are small protests.

VFinancials

Ten years of Intuit show a business that grows fast and has grown more profitable since the Mailchimp purchase compressed the margin in fiscal 2022. The figures below are for fiscal years that end on 31 July.

Revenue has compounded at 17% a year since fiscal 2017, operating income at 17%, and owner cash at 20% a year C (calculated). Outside the two deal years, fiscal 2021 and 2022, growth stayed between 11% and 16%. It is guided to 9% to 10% for fiscal 2027, which is slower than any year in the table.³ Stock pay is the number to watch. It was 6.3% of revenue in fiscal 2017 and is 9.6% now.

What an owner could really take out

This is where I part ways with the headline. Fiscal 2026 operating cash flow was $8,838M, up 42%, and reported free cash flow was $8,617M.² A buyer would think that Intuit makes $8.6B. It does not. I found Intuit through a screen for large companies whose shares had fallen 35% or more while their results held. I strip out the following.

The tax benefit. Cash taxes paid were $281M in fiscal 2026, against $1,408M in fiscal 2025 and $1,881M in fiscal 2024, even though the tax expense in the income statement was $1,451M.² The reason is a change in law. The One Big Beautiful Bill Act, signed on 4 July 2025, brought back the immediate deduction of US research and development costs, and let Intuit deduct the balance it had been spreading out. The deferred tax asset for capitalized research fell from $1,895M to $673M.² That created the $1,279M credit in operating cash flow. Intuit says plainly that its cash flow included “the impact of lower cash tax payments,” and that it expects to pay about $2B of income tax in fiscal 2027.² So the benefit will not repeat, and next year’s tax bill will be about seven times this year’s.

Fairness cuts both ways. The old rule had pushed the other way for three years: deferred taxes took $628M, $554M and $435M out of operating cash flow in fiscal 2023, 2024 and 2025, $1,617M in all.⁹ The fiscal 2026 benefit gave back most of that. Over the four years the net effect was a cost of $338M C (calculated). So the right fix is to remove deferred tax from every year, which I do in the table above. It lowers fiscal 2026, and it raises fiscal 2023 to 2025. It also changes the growth rate. Reported free cash flow grew 22% a year over the five years from fiscal 2021. On owner cash the same rate is 17% C (calculated).

Stock pay. Intuit expenses $2,056M of stock pay a year, which is not a cash outlay. But it is the price of employees’ work, and the company spends cash buying back shares to offset it. It also paid $709M in cash in fiscal 2026 to cover employee taxes on vested shares.³ I count stock pay as a cost. The old non-GAAP net income of $6,732M, which excluded it, is therefore not a figure an owner could take out.³

Lending. QuickBooks Capital lends to small businesses. Originations were $6.8B in fiscal 2026, up from $4.0B and $2.5B, and Intuit sold $2.2B of the loans.² The provision for expected credit losses, $237M, is added back in operating cash flow, but the loss is real cash. The net cash that went into loans after sales and repayments was $292M.² I take that off, since it is the capital needed to keep growing a business that is part lender. The loan book is $1,647M, funded in part by $970M of non-recourse secured borrowings, which I count in debt C (calculated).³² A lending book that grew this fast has not been through a full credit cycle.

Result: owner cash was $5.28B before lending and $5.0B after it C (calculated). That is above GAAP net income of $4.57B, because $659M of acquired-intangible amortization and $187M of depreciation are non-cash. It is 74% of the old non-GAAP net income of $6.73B, and 58% of reported free cash flow C (calculated). For fiscal 2027 the guide implies something better. GAAP operating income of $7.4B to $7.5B, taxed at 24%, with the same depreciation and amortization, less capital spending and loans, gives about $6.0B in my base E (estimate). That would be an increase of 19% on $5.0B, and it comes from the 17% job cut and 9% to 10% revenue growth. It does not come from taxes, and I expect reported operating cash flow to be lower than $8.8B next year even as earnings rise E (estimate).

Fiscal 2026 buybacks and dividends came to $6.8B, against owner cash of $5.0B. The company covered the gap with the one-time tax benefit, a cash balance that rose to $7.2B, and $1.7B of new notes.³ I would not read the $5.4B of buybacks as a sustainable run rate. It was done at an average of $410 a share, in a year when the stock fell hard. At $341 a share each dollar buys more of the company than it did, which is a point in the plan’s favour, if the plan continues.

What non-GAAP earnings mean now

Intuit reported fiscal 2026 non-GAAP earnings of $24.27 a share, against GAAP earnings of $16.46.³ The $7.81 gap is mostly stock pay ($7.42), acquired-intangible amortization ($2.38) and the restructuring charge ($1.06), less tax effects.³ From 1 August 2026 the company includes stock pay. Its fiscal 2027 non-GAAP guide is $22.88 to $23.12, which it says is 23% to 24% growth, so the fiscal 2026 figure on the new basis was about $18.60 C (calculated).³ On the new basis the stock trades at 14.8 times fiscal 2027 guided earnings and 18.3 times fiscal 2026. The old $24.27 gave 14.1 times, which flatters because it left out stock pay C (calculated). Nasdaq’s consensus for fiscal 2027 is $23.02, and its fiscal 2028 and 2029 figures are $26.30 and $29.95.²¹ The consensus does not say which basis it uses, though its level matches the guide.

Quality of earnings

Four items. First, fiscal 2026 interest and other income included $180M of gains on long-term investments, about 3% of pretax income, which is not repeatable.² Second, the effective tax rate rose to 24.1% from 20.0% because the fiscal 2025 rate was cut by $143M of excess tax benefits on stock awards, so net income growth of 18% understates the 20% growth in operating income.² Third, restructuring cost $293M in the fourth quarter and about $22M more is to come.² Fourth, the cost of revenue for QuickBooks Capital rose $115M on loan volume, which is a sign that growth in lending is not free.² None of these breaks the picture. They are why I use cash and not the EPS headline.

The balance sheet

Cash and investments are $7.2B and debt is $7.7B, so net debt is about $0.5B. Debt is $6.75B of senior notes and $0.97B of secured loans.³² The company repaid $750M of notes due September 2026 from the proceeds of $1.75B of new notes issued in June at 4.95% and 5.50%, At 31 July $1.25B was due within a year. Of that, $750M has been repaid and $500M falls due in July 2027.² It has an undrawn $2.2B revolver. Interest expense is $256M, covered more than 20 times by operating income.² Goodwill is $14.0B, 38% of assets C (calculated). That is largely the cost of Credit Karma and Mailchimp, and the 10-K says the test found fair value “substantially exceeded” carrying value for every unit in each of the last three years.² I note that Mailchimp is now guided down.

VIValuation

First, how this goes badly wrong

Before the numbers, I asked what would most likely ruin this business or this investment, and worked back. Three things could. None needs a surprise.

• The front door moves. A free AI assistant that reads a W-2, fills the form and answers questions takes the simple return, then the harder one. Intuit has already seen a version of this in its do-it-yourself line, where it “lost on price.”⁴ TurboTax made $5.3B last year.² If $1.6B of that went at a 75% margin, I put the loss at about $49 a share E (estimate). A bigger loss would reach QuickBooks and Credit Karma too.
• Price outruns customers. QuickBooks Online Accounting grew 23% on revenue per customer up 15% and paying customers up 3%.² A product can raise prices for years, and then it can’t. If Global Business Solutions, which earns 77% before platform costs, fell to the 3% customer growth rate, the whole forecast would shift.
• Capital goes wrong. Intuit spent $19.2B on two deals in about a year, and one of them is flat.⁸⁷ If a new deal of the same $12.0B lost half its value, that alone is about $22 a share E (estimate). Separately, loans of $6.8B a year have a $237M annual loss charge. If it doubled it would cost about $10 a share E (estimate).²

All three are versions of one question: is Intuit the place where a customer’s money chores live, or a product that an assistant calls when it has to? Together they set my bear case. I keep a fourth risk outside the model because I can’t size it: a federal free-filing service that returns.²

One more thing belongs here. Adobe, GoDaddy and Booking Holdings were found by the same screen as Intuit, and the idea behind all four is the same: that AI will take the customer’s first click and leave the incumbent with less. They are not four separate bets. If the idea is wrong, all four may look cheap together. If it is right, all four may be right to fall together. I treat Intuit as one view of that idea, and I don’t add the others to it.

What the price implies

At $341.09 the market values Intuit at $91.5B on 268.2M diluted shares, the 267.2M outstanding at 31 August plus 1M of dilution, taken from the gap between diluted and basic shares in fiscal 2026 C (calculated).¹²³ That is 20.7 times fiscal 2026 GAAP earnings and 16.9 times the $20.24 midpoint of fiscal 2027 GAAP guidance C (calculated).³ On the new non-GAAP basis, which includes stock pay, it is 14.8 times $23.00.³ Owner cash yield is 5.5% on fiscal 2026 and 6.5% on my fiscal 2027, against the 4.94% ten-year Treasury E (estimate).²²

What does the price pay for? Hold margin at the 31.8% of the fiscal 2027 guide, and the price needs revenue to grow 5.3% a year for ten years. Hold it at 28%, and it needs 7.2%. Or take my base growth, and the price works only if the margin in fiscal 2036 is 27%, about the 27.4% of fiscal 2026, and below the company’s guide for next year C (calculated). The market is pricing a business that adds little margin beyond this year’s, or one that grows more slowly than the guide, or a mix. That is a more cautious view than the company’s own.

What I can’t judge

Three inputs decide this valuation and I can’t know any of them. One is how fast a free assistant gets good enough for a complex return and a set of books. The second is how much of the 17% job cut stays saved once Intuit has to compete on price. The third is the quality of a loan book that has grown threefold in three years and has not met a recession. I can read the numbers on each, and I can’t read what happens next. So here is the range, not one figure. Read my $428 with the grid beside it.

The discount rate

I use 10.0%. The ten-year Treasury yields 4.94%, and Damodaran’s implied equity risk premium was 4.20% on 1 October.²²²³ My ten-year beta on month-end returns against the S&P 500 is 1.03, which gives 9.3%. The five-year beta of 1.01 gives 9.2%, and the three-year beta of 0.59 gives 7.4% C (calculated).¹²⁴ The three betas differ a lot because the stock has moved in and out of step with the market. I round up to 10.0%, a little above what the ten-year beta gives, because my cash flows are estimates. It matches the rate on my ServiceNow page, is a little under the 10.5% on Broadcom, and is above the 9.0% on Visa.

Three cases

The model is simple on purpose. For each year from fiscal 2027 to 2036 I set revenue growth and GAAP operating margin. I tax operating income at 24%, the rate in the 10-K. I add back depreciation, at 0.9% of revenue, and the amortization of acquired intangibles, which is non-cash and falls from $0.66B to $0.28B. Then I take off capital spending and capitalized software at 1.0% of revenue, and the net cash that goes into loans, $0.30B in fiscal 2027 and growing with revenue. Stock pay is already inside operating income, so I don’t add it back. What is left is owner cash.

The fiscal 2027 figures follow the guide: revenue of $23.4B and operating income of $7.45B, against ranges of $23.28B to $23.51B and $7.41B to $7.49B.³ My base earnings, put on the company’s new non-GAAP basis, come to about $26 a share in fiscal 2028 and $28 in fiscal 2029, before any buybacks E (estimate). The Nasdaq consensus is $26.30 and $29.95, which include some share reduction.²¹ So my base is about the consensus, and the near years are not where I differ from the market. The question is what happens after that, and whether the 10% discount is fair.

Where would $30B of fiscal 2030 revenue come from? Mostly from the parts management calls Big Bets. Global Business Solutions, growing about 10.5% a year, would be about $19B, and Consumer, growing about 4.5% a year, about $10B E (estimate). Management’s own three-year targets, from a call summary, are 10% to 15% for Global Business Solutions and 4% to 8% for Consumer.¹² My base sits at the low end of both. That is a deliberate choice: I want the base to be what Intuit says it can do, minus a haircut, not above it.

The price sits 20% below my base, and that is the plain result. But why might the market be right? Because it can see what I can. TurboTax lost price-sensitive filers. Units fell. The customer count in the 10-K is lower. Growth is leaning on price. A cut of 17% of staff came with the report on a weak tax season, following another cut two years before, and two shareholder suits say that the company misled investors.² Stocks do not fall 58% for no reason. The market may be saying that Intuit is a very good business that is about to have its front door reorganised. If so, my growth is too high and the margin too, and the grid shows the cost. The $226 bear is a fair picture of that world. The model’s value also depends on years beyond fiscal 2036: 55% of my base enterprise value comes after it C (calculated). If a reader weighs the far years less, the value is lower.

Other uses of the same dollar

The ten-year Treasury pays 4.94%. Against that, my base here returns 11.7% a year and my bear 7.5% E (estimate).²² The closest comparison on this site is ServiceNow, because it is also software priced for the fear of AI agents, and there my base cash flows return about 11% a year at $105. Mastercard, a steadier business, returns 8.4% on my base. Those pages use their own prices, rates and methods, so I treat gaps of a point as noise. Intuit’s base return is at the high end of that group, and the reason is the same one the market gives: the fear.

It also helps to remember what the screen is. Adobe, GoDaddy and Booking share this idea with Intuit, and so a purchase of several of them is closer to one position than four. Their pages here use their own prices and methods, so I don’t set their returns beside mine.

What past lows looked like

The stock has fallen more than 30% from a record close three times in ten years: 36% in 2020, 49% from November 2021 to May 2022, and 68% to the low close of $255.07 on 25 June 2026 C (calculated).¹ At the 2022 low of $353.31, owner cash of $2.2B in fiscal 2022 was a yield of about 2.2% on the market value then C (calculated). At the June 2026 low, fiscal 2026 owner cash after lending was a yield of about 7.3% C (calculated). The stock was far more expensive at the 2022 low than at this one. That is not a forecast. It shows that investors were paying much more for the same franchise four years ago, before the AI question.

The margin of safety

• Price$341.09Close, Q4 2026 [1][25]
• Base value$428My estimate E (estimate)
• Margin of safety below$301set from my bear case C (calculated)

I set the margin of safety from the bear case: at $301, my $226 bear value is three quarters of the price paid, and the price is 30% under my base value. I use the bear because the danger here is the shape of the business changing, and a discount to base would not protect against that. At $301 my bear returns 8.1% and my base 12.8% a year E (estimate). At $341.09 the price is 13% above that level, so there is some cushion against my base and no full margin of safety. The stock closed at $255.07 in June, under it. The margin of safety does not cover the stress case either: revenue growth falling to a decline, margin at 21%, loans at 1.3 times my base, which gives $167.

Two different questions

How good is the business, and how good is the price? They are separate questions. As a business Intuit is very good on most tests I have: two segments that earn above 70% before platform costs, little net debt, cash that comes from customers and not from financing, and a ten-year record of growth near 17% a year. Its flaws are in the details: stock pay near 10% of revenue, a growth rate that depends on price, a weak purchase in Mailchimp, and a front door that is exposed. As for the price, it was poor at $807. My base gave only 6.8% a year there. At $341.09 it looks reasonable on my numbers, with one question still open. So this looks like a good business at a fair-to-good price with an unresolved question, and the question is one the market has not answered either. The price reflects a debate, and I can’t settle it from here.

Where this could be in 5 and 10 years

The value at each date is my model’s value of the owner cash after it, at 10.0%, less $0.47B of net debt, per today’s 268.2M shares and in that year’s dollars. Owner cash is the model’s cash after stock pay and loans, for fiscal 2027 to 2036. The return counts price growth and that cash, exiting at my July 2036 value E (estimate).

The base holds at five years if revenue reaches $31.6B from $21.4B, the margin reaches 35%, and owner cash grows from $5.0B to $8.5B a year. It holds at ten if revenue reaches $41B at a 36% margin, with stock pay kept near 8% to 9% of revenue. Value rises only 3% to 4% a year at 10%. Most of the 13.4% is the $323 a share paid out. It is higher than the 11.7% earlier because it values cash after 2036 at 10%, not at the rate that matches today’s price.

The bear is a squeeze. Growth falls to zero by FY31, margin slips to 24.5%, and owner cash stays near $5B, so value barely moves. The bull is the reverse: growth near 10% to FY31, margin 42% by FY36 and revenue of $53B.

The most likely way my base is wrong is growth. Intuit guides to its slowest year in ten, and price carries much of it. If growth runs 2 points a year below my base, the 2036 value is $497, not $598, and the return falls from 13.4% to 11.3%. A margin 3 points lower gives $546. An 11% discount rate gives $523. I can’t judge how fast a free assistant becomes good enough for a complex return, so I don’t put one number on it.

These are estimates of value, not forecasts of the share price. The market price can sit above or below value for years.

VIIThe case for the price

What most people miss

Most of the fall in TurboTax units is free filers: about 7M paid nothing, down from 8M, and paying online units were expected to rise 2%.¹¹ The do-it-yourself market where Intuit says it lost on price is $5B, against $37B for assisted tax.⁴ The fiscal 2027 guide adds about $1.3B of operating profit on $1.9B of new revenue once the restructuring charge is removed, two thirds of each new dollar C (calculated).³

The bull case

AI makes the work cheaper, and the firm that owns the data, the license and the liability gains. Intuit has the returns, the books of about 10 million businesses and about 1 million accountants, and it is cutting costs while it grows.⁴ On that view the fall from $807 is a mistake about who wins. A call summary gives management three-year targets of 10% to 15% for Global Business Solutions and 4% to 8% for Consumer.¹² Yahoo Finance reported that Morgan Stanley named Intuit a top pick on 19 March 2026, with a $580 target. That was two months before the 20% fall, and I don’t know if the view stands.²⁶

The funds I can check left. Fundsmith, Polen Capital and ARK held shares in late 2025 and none by June 2026, and fourteen others, among them Altimeter, Berkshire Hathaway and Coatue, held none.^27282930 Filings show positions, not reasons. I found nothing on Intuit from Gerstner, Baker, Wood, Gurley, Chamath, Sacks or Ackman.

Against reality

Much of the case is in the guide: $23.4B of revenue at a 31.8% margin for fiscal 2027, which my base takes.³ After that the base needs growth of 8.5%, 8.0% and 7.5% in FY28 to FY30, and a 33.9% margin by FY29 E (estimate). If Intuit hits its guide, holds 8% to 9% growth for two more years and keeps the savings from the job cut, $428 is possible and the price sits below it. The weak points are guided growth, the slowest in my ten years, and a margin step that rests on a staff cut a few months old.³¹³

Multiplied out, my base needs five things: growth of 8% or more through FY29 (65%), cuts that stick (70%), TurboTax revenue that doesn’t fall (80%), no value-destroying deal (85%) and loan losses near 3.5% of originations (85%). My guesses multiply to 26%, perhaps 30% with overlap E (estimate). The price needs less: about 5% growth, a 31.8% margin and TurboTax holding, say 80%, 75% and 85%, or 51% E (estimate). So the price can ask for less than my base without my base being likely.

What a 10x would take. Ten times $341.09 is $3,411, about $915B of market value. At 25 times earnings that is about $37B of net income, eight times fiscal 2026’s $4.6B. At a 30% net margin, against 21% in fiscal 2026, it is about $122B of revenue, 5.7 times today’s, or 19% growth a year for ten years from a company guided to 9% to 10% C (calculated). My bull case reaches $53B and $593. Intuit’s own ten years gave 3.1 times from $108.74 on 31 October 2016, and its best was 7.4 times to the July 2025 high C (calculated).¹ The one tenfold rise in my ten-year file is NVIDIA, from about $257B in July 2020 to about $2.6T in May 2024, on a surge in chip demand C (calculated).³¹³² I can’t show an application software company that has done it. For Intuit, tenfold from here is close to impossible, and I would not build a case on it.

Where I come down

The tax benefit and the loans moved me most. Removing the one-time benefit and the $292M that went into loans takes owner cash from $8.6B of reported free cash flow to $5.0B, so I start from the guide and not from a constant growth rate. “We lost on price” moved me toward a lower bear. What would raise my values most is a year in which TurboTax revenue grows after the price reset, with business customers growing faster than 3%.

VIIIRisks, and what would change my view

The three ways this goes badly wrong are at the start of Valuation. These are the rest.

• Free federal filing. The IRS system was suspended, and the 10-K says proponents still push for it.² If a direct filing program returned, it would meet the simplest filers first.
• Price rises and churn. A product that grows through price eventually meets customers who leave. Intuit does not report churn, so a rise would show up late in paying-customer growth.²
• Credit Karma and lenders. Credit Karma earns when lenders approve customers. In a credit downturn its revenue could fall faster than costs.²
• Lending. Intuit lends to small businesses and holds a $1.6B book, funded in part by non-recourse debt.² The book has grown fast, and a recession is the test.
• People and execution. Two staff cuts in three years, a 17% cut that came days after a weak season, the head of the Small Business Group gone, and a new general counsel.¹³¹⁶² Cuts of this size can help, and they can also remove people who know how the product works.
• Legal. Two shareholder class actions allege misleading statements. A class action in Ontario was certified in July 2026. The FTC order was vacated in March, and a motion to dismiss the last FTC suit as moot was filed on 31 August 2026.² The company says it cannot estimate losses.
• Data and security. Intuit holds tax returns and bank links. An FTC order requires a security program for Credit Karma data through 2034.² A breach would hit trust, which is what customers pay for.
• Stock pay. It is near 10% of revenue. The new non-GAAP definition helps, and the cost is still large.³